Trust and Scandal: a Tale of Two Theories

Sigma: Journal of Political and International Studies

Volume 32 Article 7

2015

Trust and Scandal: A Tale of Two Theories

Follow this and additional works at: https://scholarsarchive.byu.edu/sigma

Part of the International and Area Studies Commons, and the Political Science Commons

Recommended Citation (2015) "Trust and Scandal: A Tale of Two Theories," Sigma: Journal of Political and International Studies: Vol. 32 , Article 7. Available at: https://scholarsarchive.byu.edu/sigma/vol32/iss1/7

This Article is brought to you for free and open access by the Journals at BYU ScholarsArchive. It has been accepted for inclusion in Sigma: Journal of Political and International Studies by an authorized editor of BYU ScholarsArchive. For more information, please contact [email protected], [email protected]. Trust and Scandal: A Tale of Two Theories by James Martherus

“I am not a crook.” Those five words are burned into the nation’s memory. Pres- ident Nixon was under investigation after five men were arrested trying to break into the Watergate office complex, where the Democratic National Committee was headquartered. Evidence was found that the burglary was funded using money from Nixon’s re-election campaign. This led to cover-ups at high levels within the FBI, CIA, and the White House. Nixon was implicated in these cover-ups and was eventually charged with obstruction of justice, abuse of authority, and defiance of committee subpoenas. It was then that Nixon uttered the infamous phrase, “I am not a crook.” He was not the first public official to be involved in a major scandal, andhe would not be the last. Just over twenty years later, President Clinton was involved in at least one affair, with intern Monica Lewinsky. Again, the nation was rocked as Clinton was impeached on charges of perjury and obstruction of justice. The State of Utah has recently experienced its own scandal. Attorney General John Swallow took office in January 2013 and almost immediately came under investigation by the FBI (Gehrke 2013a). The FBI investigated whether Swallow had con- spired to help an indicted businessman, Jeremy Johnson, avoid a lawsuit by the FTC (Carlton 2013). The investigation did not lead to any charges against Swallow, but it did provoke further inquiry by the lieutenant governor’s office and the Utah House (Gehrke 2013b; HR9001) about various alleged misconducts including violation of campaign laws. Swallow resigned as Utah’s attorney general in November of that year (Roche and Romboy 2013) and was arrested on various state charges (Crofts 2014). Swallow maintains his innocence, but whether the accusations are true or not, Utah has been rocked by the news. These events provide a unique opportunity to study the effects

28 SIGMA of political scandal on public opinion. Specifically, I present two findings. First, political scandals negatively affect trust in politicians but not in government institutions. Sec- ond, demographics like religion and party identification affect perception of the guilt of the parties involved, because individuals are less likely to believe someone like them could be involved in a scandal. Theory In the 1960s and early 70s, the public trusted the government at levels that seem laughable today. Pew Research found in 1964 that 77 percent of respondents felt they could trust the government “always” or “most of the time.” Since 1964, trust in government has experienced a steady decline, reaching a low of 20 percent in 2014 (Pew 2014). This decline in trust is not unique to the U.S., but a phenomenon observed in virtually all Western democracies (Dalton 2004; Putnam et al. 2000). Political scientists have recently tried to explain this phenomenon. The literature proposes many theories, most of which focus on the perceptions that government wastes too much, is inefficient, or chooses to spend on the wrong things (Nye et al. 1997; Chanley et al. 2000; Keele 2007; Miller and Borrelli 1991). A smaller group of researchers have studied the effects of political “scandals” on public trust in government. Two theories have emerged as a result of this research: one theory suggests that scandals can negatively affect public confidence in government institutions (i.e., Congress, the Supreme Court, etc.), while the other suggests that scandals have a large effect on confidence in the individuals involved in them but not on the institutions themselves. Shaun Bowler and Jeffrey Karp support the first theory, suggesting that scandals have an effect on the public’s attitudes toward institutions and the political process (2004). They are not alone; many other studies have suggested that corruption can have negative repercussions, not only for public confidence and trust but also for the outcome of elections (see e.g., Anderson and Tverdova 2003; Banducci and Karp 1994; Clausen et al. 2011). Other academics have questioned whether scandal really affects the perception of institutions and find that scandals only affect the individual politicians involved (Clarke et al. 1998; Lanoue and Headrick 1994). For example, Jürgen Maier found that political scandals contribute to a decline of political support for candidates but not a loss of confidence in government institutions (2011). Diminished trust in individual politicians has been observed after Watergate (Yaffee 1999), Iran-Contra (Krosnick and Kinder 1990), and other major scandals, with the notable exception of the Clinton- Lewinsky scandal (Zaller 1998). The idea that scandals do not really affect trust in the democratic process and government institutions has grown more and more popular in recent years. One of the most recent studies in this area attempts to explain this by mak- ing a counterintuitive claim: Scandals may have had a larger impact on public trust in institutions at one point, but the growing frequency of political scandals has caused us to be less affected by them (Kumlin and Esaiasson 2012).

29 MARTHERUS

Some research suggests that even for the individuals involved, scandals do not have a huge effect. For example, Stephen Bennet found that during the Clinton- Lewinsky scandal, approval ratings for President Clinton suffered an initial drop but then stayed relatively high. Further, most individuals did not believe the scandal was serious enough to warrant his removal from office (2002; see also Zaller 1998). I am interested in further explaining the effect of scandal on public trust in government officials and institutions. Specifically, I am interested in studying which of the two above theories are more correct. The current literature fails to account for important variables like religion. Utah is unique in that most of its inhabitants are members of one church: The Church of Jesus Christ of Latter-day Saints. Recently, Utah’s Attorney General John Swallow was involved in a political scandal, where he was accused of accepting bribes, tampering with evidence, and misusing public funds. Swallow also held a leadership position in the Church, providing a unique opportunity to study how religion affects public confidence in government when a scandal has occurred. I asked whether participants have heard of the John Swallow scandal and then measure the levels of public confidence among those who have followed the scandal and those who have not. I also controlled for participants’ religion and activity in that religion to see if religion has an effect on public perception of scandals involving those of their own faith. As a secondary research goal, I plan to study how different demographic factors affect how we view government scandals. Many people tend to trust religious individuals more than nonreligious individuals (Tan 2008). Furthermore, people tend to feel more warmly about people who share their religion than about people from other religions (Putnam 2010). When we trust someone more deeply, a breach of that trust is more devastating. Therefore, those who share John Swallow’s religion will be more likely to lose trust in politicians/government in general after hearing that John Swallow, a prominent member of their church, may have participated in corrupt activities than would an individual who does not share a religion with John Swallow. Further, we tend to trust those who are more like us politically (Lambert et al. 1986), and we find it hard to believe that “our team” would do something unethical. Therefore, those who identify as Republican will be less likely to believe that John Swal- low is guilty than those who identify as Democrats. Considering the above, I have two hypotheses of interest. First, I expect to find that a government scandal will cause individuals to lose trust in individual politicians but not in government institutions. Second, I expect to find that those who are similar to John Swallow (religion, ideology, race, etc.) are less likely to believe that he is guilty of corruption. Methods The data used in this analysis were gathered during the Utah Colleges Exit Poll (hereafter UCEP). The UCEP is conducted biannually on Election Day. About six hundred undergraduate students volunteer as pollsters and are sent to over one

30 SIGMA hundred polling locations throughout the state of Utah. The sample for the poll is drawn by statistics majors and uses a stratified, multimode design. The sample contains four strata, one for each congressional district. Within each stratum, we clustered on voting districts and polling places. In addition to the pollsters located at polling locations, we also collected data from mail-in and early voters through online and telephone surveys. The poll accurately predicted Utah’s major races. For example, in the fourth congressional district, Mia Love garnered 50.0% of the vote, while Doug Owens received 46.8% of the vote. The UCEP called the race for Mia Love with 50.6% to Doug Owens’ 47.0%. All of the survey questions used in this paper are replicated from either a reputable survey research firm (i.e., Pew research) or the Utah Voter Poll. The Utah Voter Poll has a long history of accurate survey results in Utah. The Appendix includes the full text of all questions used in this paper. Using the data from the UCEP, I ran various statistical tests. I first used T-tests and crosstabs to conduct preliminary analysis and then ran simple linear regressions. All regressions were run using robust standard errors to account for possible heteroskedasticity and using proper weighting techniques. I dropped observations that had been marked invalid or missing as well as the “don’t know” responses, which were analyzed separately. My models included various control variables such as age, gender, race, party identification, and ideology. Some models also included a knowledge index. The knowledge index was constructed by asking four political knowledge questions and adding up the number of questions the respondent answered correctly. The purpose of the knowledge index was to gauge a respondent’s general political interest. Presum- ably, individuals who follow politics more closely would be more likely to be affected by a government scandal. Analysis/Results The first thing to note about our results is John Swallow’s favorability rating. Respondents were asked to rate various politicians on a scale from 1 to 5. Consistent with what we would expect given the literature (see e.g., McAllister 2000; Hetherington 1999), John Swallow’s favorability rating was incredibly low, with over 54 percent rating him at only 1 out of 5 and an average score of about 1.5. Compare this with the way Utah voters feel about President Barack Obama, who is hardly a popular figure in Utah. The president received an average score of 2.3, meaning these voters hold even more negative views about Swallow than they do about the president. This low approval rating is likely related to the fact that almost all voters had heard something about Swallow’s scandal. More than 90 percent of Utah voters claim to have heard, read, or seen something about a scandal involving the former attorney general. Unfortunately, the high level of awareness on this issue made it difficult to prove my hypotheses, because there were not enough respondents who had not heard about the scandal to produce a satisfactory level of variance. I now discuss each of my hypotheses in detail.

31 MARTHERUS

My first hypothesis was that government scandals would cause a loss of trust in the politicians involved, but not in government institutions. My analysis largely supports this hypothesis. To measure the loss of trust in Swallow, I used his favorability rating. Those who had not heard of the scandal gave him an average of 2.6 on the favorability scale, while those who had heard about the scandal gave him an average rating of 1.6. Regression analysis produces this same finding, even controlling for religion, race, party identification, and ideology. Table one shows the results of this regression. This regression clearly shows that having knowledge of a government scandal negatively impacts how the public views the individuals involved. Although not part

Table 1: Effect of scandal on John Swallow’s favorability Dependent variable: Favorability rating from 1 to 5 Regressor Heard about scandal -.94** (.129) Religion Index -.04* (.018) White -.28** (.094) Republican .16** (.040) Liberal -.07** (.026) Female .02 (.042) Income -.02 (.011) Married -.02 (.054) Education Level -.03 (.024) Intercept 2.82** (.254) Regression summary statistics – R2 .12 R2 .12 SER .90 n 2927 Notes: Dependent variable is the respondents rating of John Swallow on a scale from 1 to 5 (see appendix for full question wording). Religion index refers to an index plotting religious preference and activity in that religion. Heteroskedasticity-robust standard errors are given in parentheses under estimated coefficients. Coefficients are significant at the *5%, **1% significance level.

32 SIGMA of my research, it is also interesting to note which of the demographic factors affected Swallow’s favorability. Some factors, like party identification and ideology, we would expect. Others, like race and religion, are more interesting. The other part of my first hypothesis was that scandals would not significantly affect levels of trust in government institutions. The data I collected mostly supports this part of the hypothesis. We asked respondents two questions. The first asked how much they trusted the federal, state, and local government to do the right thing. The second asked them to rate their level of confidence in various Utah government offices, including the office of the attorney general. My hypothesis would suggest that knowledge of the Swallow scandal should not significantly affect levels of trust or confidence in these two questions. Tables two and three show the results of these two questions. This first set of regressions is inconclusive as far as my hypothesis is concerned. Surprisingly, knowledge of the scandal had no significant effect on levels of confidence in the attorney general’s office but did have quite a large effect onlevels

Table 2: Effect of scandal on confidence in government offices Dependent variable: Confidence in government offices Regressor Office of the Office of the Utah Attorney Utah Gover- General nor Heard about scandal -.08 .20** (.070) (.068) Religion Index -.06** -.08** (.018) (.014) Female -.0005 -.04 (.037) (.032) Republican .13** .29** (.033) (.033) Liberal -.07** -.06** (.026) (.022) Intercept 2.64** 3.15** (.162) (.160) Regression summary statistics – R2 .09 .24 R2 .09 .28 SER .81 .73 n 3277 3346 Notes: Dependent variable is confidence in various government offices (name of office at top of each column) on a five-point scale. Religion index refers to an index plotting religious preference and activity in that religion. Heteroskedasticity-robust standard errors are given in parentheses under estimated coefficients. Coefficients are significant at the *5%, **1% significance level.

33 MARTHERUS of confidence in the Utah governor’s office. Those who had followed thescandal actually rated the governor’s office higher than those who had not. Possibly, those who had followed the scandal approved of the way the governor’s office had dealt with the scandal by launching an extensive investigation. Other factors that affected confidence levels included religion, party identification, and ideology. This is unsurprising; Latter-day Saints (LDS) conservative Republicans control Utah, so LDS conservative Republicans are more likely to have confidence in Utah’s government. This second set of regressions gives substantial evidence against the theory that government scandals cause a loss of trust in government institutions. For all

Table 3: Effect of religion (and other demographics) on trust in different levels of government Dependent variable: How often do you think you can you trust the following organizations to do what is right? Regressor Federal State Local Government Government Government Heard about scandal -.06 .01 -.05 (.040) (.046) (.051) Religion Index .01 -.04** -.05** (.0093) (.012) (.013) Female -.06** -.12** -.13** (.022) (.028) (.030) Republican -.15** .09** .04 (.022) (.025) (.026) Liberal .12** -.04** -.02 (.015) (.017) (.019) White -.12* .02 -.03 (.042) (.051 (.062) Knowledge Index -.08 -.16* -.03 (.053) (.067) (.081) Intercept 1.59** 2.15** 2.12** (.115) (.134) (.158) Regression summary statistics – R2 .25 .08 .04 R2 .25 .09 .04 SER .49 .58 .63 n 3332 2586 3347 Notes: Dependent variable is how often you think you can trust the given organizations. Full question text given in the appendix. Religion index refers to an index plotting religious preference and activity in that religion. Knowledge Index refers to an index of four political knowledge questions where knowledge index is equal to the percentage of questions they answered correctly. Heteroskedasticity-robust standard errors are given in parentheses under estimated coefficients. Coefficients are significant at the *5%, **1% significance level.

34 SIGMA three levels of government, knowledge of the Swallow scandal had no statistically significant effect on levels of trust. Interestingly, the factors that do affect levels of trust vary across the levels of government. For example, religion affected levels of trust in the state government but not the federal government. Local government seems to be a unique case. Surprisingly, neither party identification nor ideology had an effect on an individual’s trust in local government. This can probably be attributed to the fact that local governments tend to enjoy high levels of trust and satisfaction or the fact that many respondents probably know some of the people in their local government personally. Overall, my analysis of the data supports the hypothesis that government scandals negatively affect the politicians involved in them but not the government in general.

Table 4: Effect of religion (and other demographics) on whether or not an individual believes John Swallow has done something illegal or unethical Dependent variable: John Swallow has done something illegal, unethical, or neither Regressor Religion Index -.06* (.028) Female .18** (.065) Age -.002 (.002) Republican .17** (.062) Liberal -.12** (.044) White -.28* (.139) Education Level -.03 (.035) Married .11 (.088) Income Level -.05** (.016) Regression summary statistics Pseudo R2 .06 Wald’s Chi2 116.30 n 2685 Notes: Dependent variable is whether the respondent thought John Swallow had done something illegal, unethical, or neither (see appendix for full question wording). Religion index refers to an index plotting religious preference and activity in that religion. Standard errors are given in parentheses under estimated coefficients. Coefficients are significant at the *5%, **1% significance level.

35 MARTHERUS

My second hypothesis was that those who are more similar to Swallow demo- graphically would be more likely to believe that he is innocent of the charges brought against him. The analysis partially supports this hypothesis. Data analysis reveals that some demographics matter and some do not. Gender and religion, for example, seem to make a significant difference. Age, education level, and marital status, however, had no significant effect. Table four displays the results of this analysis. There are a few things that should be noted about these results. As the Pseudo R2 measure indicates, this model is not very predictive. This is unsurprising given that there are myriad other considerations one may have when deciding whether they think John Swallow is guilty. In order to make a more predictive model, we would need to collect additional information from these respondents. Another point of interest is religion. Active LDS tend to be more likely to believe Swallow is innocent. Perhaps this is because Swallow was an active LDS and served as a bishop for a time. In conclusion, my second hypothesis was partially correct. There is some evidence to support the idea that we tend to trust those who are similar to us more than those who are different. However, some similarities like education level or marital status seem to have no effect. Further exploration of this phenomenon is necessary, but I propose one theory as to why this might be the case. Some demographics tend to measure similarities that are based on identity. We identify as a democrat or republican and feel a sort of kinship with those who identify as we do. The same phenomenon occurs within religion, perhaps even more so. However, we tend to not feel a strong bond with someone just because they happen to be married or to have the same level of education as us. Therefore, these types of identifiers understandably have little effect on how much we trust someone. Limitations While the Utah Colleges Exit Poll was a considerably accurate poll, there are still some limitations to its use as a representative sample of Utah voters. It is well known, for example, that midterm election voters are significantly different from general election voters. Midterm election voters tend to be older, whiter, more educated, and more partisan (Wattenberg and Brians 2002). Our results are weighted more heavily toward those groups. Another limitation to using this data is the low number of respondents who responded that they had not heard about accusations involving Swallow. We had planned on doing much of our analysis between those who had heard of the scandal and those who had not; however, over 90 percent of voters claimed to have heard of the scandal. One possible explanation for this extremely high percentage is that voters in midterm elections tend to be more politically active than those in general elections. There may be an even stronger bias toward informed voters in Utah, as voter turnout was a mere 30 percent. This would suggest that only the most dedicated and careful voters actually voted. Another possible explanation is that respondents did

36 SIGMA not want to acknowledge ignorance. Essentially, because so few respondents claimed to have not heard about the Swallow scandal, we could not effectively compare the two groups. The Utah Colleges Exit Poll also suffers from some nonresponse error. Student pollsters are instructed to gather as much data as possible from nonrespondents (i.e., race, approximate age, gender) in order to check for nonresponse error. We have weighted the results to reflect any error, but some small error is inevitable. There is some evidence to suggest that asking someone if they have heard about the Swallow scandal is not a perfect substitute for getting a respondent’s opinion before and after a scandal has occurred. It is possible that people who said they had heard about the scandal really had not and vice versa. For example, I conducted a t-test to see if the mean confidence level in the Utah attorney general’s office was lower than the mean confidence level in Utah’s Child Protective Services (CPS). In theory, CPS may be used as a control since they have not been involved in any major scandals recently and they are relatively unknown like the attorney general’s office. The t-test found a statistically significant difference between the two means, with the attorney general’s office scoring lower than CPS. However, the difference was only a tenth of a point, and I do not believe this difference alone constitutes enough evidence to disprove the hypothesis that scandals affect the trust in the individual rather than the office itself. Finally, while the question wording we used was a replication of questions used in Gallup and Pew research studies, further study suggests there are better methods for asking about trust in government. Specifically, Timothy Cooke and Paul Gronke have developed a measure they claim is much more accurate than those used in most studies. They suggest using a scale from zero to ten where zero is very strong distrust of government to do the right thing, ten is very strong trust in government to do the right thing, and five is neither trust nor distrust in the government (2005). In future research, I will use these measures to more accurately describe trust in government among respondents. Conclusion Political scandals have been around for a long time and show no signs of ceasing. In fact, as the world advances technologically, our ability to uncover these scandals may increase, exposing us to scandals and corruption with even greater frequency. In light of this possibility, perhaps it is comforting to know that scandals do not sub- stantially affect the government’s ability to function or the people’s general trust in that ability. When we uncover scandals, we can throw those involved out of office and try again. On the other hand, could a higher volume of government scandals begin to erode the public trust? This paper has studied the effects of a single scandal but may not be applicable to locations that experience scandals fairly regularly. Further research should be done to determine the effects of multiple scandals.

37 MARTHERUS

REFERENCES Anderson, Christopher J., and Yuliya V. Tverdova. 2003. Corruption, political allegiances, and attitudes toward government in contemporary democracies. American Journal of Political Science 47, no. 1 (January): 91–109. Banducci, Susan A., and Jeffrey A. Karp. 1994. The electoral consequences of scandal and reap- portionment in the 1992 House elections. American Politics Quarterly 22 (January): 3–26. Bennet, Stephen Earl. 2002. Another lesson about public opinion during the Clinton-Lewinsky scandal. Presidential Studies Quarterly 32, no. 2 (June): 276–92. Bowler, Shaun and Jeffrey A. Karp. 2004. Politicians, scandals, and trust in government. Political Behavior 26, no. 3 (September): 271–87. Carlton, Jim. 2013. Utah official fights allegations of briber. The Wall Street Journal January 22. Chanley, Virginia A., Thomas J. Rudolph, and Wendy M. Rahn. 2000. The origins and consequences of public trust in government. The Public Opinion Quarterly 64, no. 3 (Autumn): 239–56. Citrin, Jack. 1974. Comment: the political relevance of trust in Government. The American Jour- nal of Political Science 68, no. 3 (September): 973–88. Clark, Harold D., Marianne C. Stewart, and Paul F. Whitely. 1998. New models for new labour: the political economy of labour party support. The American Political Science Review 92, no. 3 (September): 559–75. Clausen, Bianca, Aart Kraay, and Zsolt Nyiri. 2011. Corruption and confidence in public institutions: Evidence from a global survey. The World Bank Economic Review 25, no. 2: 212–49. Cooke, Timothy E., and Paul Gronke. 2005. The Skeptical American: Revisiting the Meanings of Trust in Government and Confidence in Institutions. The Journal of Politics 67, no. 3 (August): 784–803. Crigler, Ann. 2000. Leadership image-building: After Clinton and Watergate. Political Psychol- ogy 21, no. 1 (March): 179–98. Crofts, Natalie. 2014. Swallow, Shurtleff taken into custody. KSL July 15th. Dalton, Russel J. 2004. Democratic Challenges, Democratic Choices: The Erosion of Political Sup- port in Advanced Industrial Democracies. Oxford: Oxford University Press. Damico, Alfonso J., M. Margaret Conway, and Sandra Bowman Damico. 2000. Patterns of political trust and mistrust: Three moments in the lives of democratic citizens. Polity (Spring): 377–400. Doherty, David, Conor M. Dowling and Michael G. Miller. 2011. Are financial or moral scandals worse? It depends. Political Science and Politics 44, no. 4 (October): 749–57. Gehrke, Robert. 2013a. Utah Attorney General John Swallow under investigation, feds confirm. Salt Lake Tribune January 25th. Gehrke, Robert. 2013b. Special council will investigate John Swallow’s financial arrangements. Salt Lake Tribune August 8th. Hetherington, Marc J. 1998. The political relevance of political trust. The American Political Sci- ence Review 92, no. 4 (December): 791–808. HR9001. http://le.utah.gov/~2013H1/bills/hbillenr/HR9001.htm. Jones, Jeffrey M. 2014. Americans’ trust in Executive, Legislative branches down. Gallup Politics. September 15. http://www.gallup.com/poll/175790/ americans-trust-executive-legislative-branches-down.aspx. (Accessed September 19, 2014). Keele, Luke. 2007. Social capital and the dynamics of trust in government. American Journal of Political Science 51, no. 2 (April): 241–54. Krosnick, Jon A., and Donald R. Kinder. 1990. Altering the foundations of support for the Presi- dent through priming. The American Political Science Review 84, no. 2 (June): 497–512. Kumlin, Staffan, and Peter Esaiasson. 2012. Scandal Fatigue? Scandal elections and satisfaction with democracy in Western Europe, 1977-2007. British Journal of Political Science 42, no. 2 (April): 263–82.

38 SIGMA

Lambert, Ronald D., James E. Curtis, Steven D. Brown, and Barry J. Kay. 1986. Effects of identification with governing parties on feelings of political efficacy and trust. Canadian Journal of Political Science 19, no. 4 (December): 705–28. Lanoue, David J., and Barbara Headrick. 1994. Prime ministers, parties, and the public: the dynamics of party popularity in Great Britain. Public Opinion Quarterly 58 no. 2 : 191–209. Maier, Jürgen. 2011. The impact of political scandals on political support: An experimental test of two theories. International Political Science Review 32, no. 3 (June): 283–302. McAllister, Ian. 2000. Keeping them honest: public and elite perception of ethical conduct among Australian legislators. Political Studies 48, no. 1 (March): 22–37. Miller, Arthur. 1974. Political issues and trust in Government: 1964-1970. The American Political Science Review 68, no. 3 (March): 951–72. Nye, Joseph S., Philip Zelikow, and David C. King. 1997. Why People Don’t Trust Government. Harvard University Press. Putnam, Robert D and David E. Campbell. 2010. American Grace: How Religion Divides and Unites Us. Simon & Schuster NY: New York. Pg. 508. Putnam Robert D, Pharr SJ and Russel J. Dalton. 2000. Introduction: what’s troubling the trilateral democracies? In: Pharr SJ and Putnam RD (eds) Disaffected Democracies: What’s Troubling the Trilateral Countries? Princeton, NJ: Princeton University Press. Roche, Lisa Riley, and Dennis Romboy. 2013. Utah A.G. John Swallow resigns, citing toll on fam- ily, finances. Deseret News November 21st. Salt Lake Tribune. Swallow, Shurtleff arrested, face 23 counts, up to 30 years in prison. July 19th, 2014. Tan, Jonathan H.W. 2008. Religion and trust: An experimental study. Journal of Economic Psy- chology 29, no. 6 (December): 832–48. Wattenberg, Martin P., and Craig Leonard Brians. Partisan turnout bias in midterm legislative elections. Legislative Studies Quarterly 27, no. 3 (August): 407–21. Yaffee, Robert A. 1999. Presidential scandals and job approval: Impact analysis with SAS. Avail- able at: http://www.nyu.edu/its/socsci/docs/scandalshtml (accessed November 2014). Zaller, John R. 1998. Monica Lewinsky’s contribution to political science. Political Science and Politics 31, no. 2 (June): 182–89.

APPENDIX Full Text of Survey Questions Have you heard, read, or seen anything about accusations involving former Attorney General John Swallow? Yes No

How much of the time do you think you can trust the following organizations to do what is right? (Only some of the time, most of the time, just about always, don’t know) Deseret News The Salt Lake Tribune Federal Government State Government Local Government

How much confidence do you have in each of the following government offices? (Not at all, Not very much, Some, A great deal, Don’t know) Office of the Utah Attorney General Office of the Utah Governor Child Protective Services

39 MARTHERUS

As you may know, the media have reported several accusations about the activities of former Attorney General John Swallow. Based on what you have heard or read, which of the following statements best described your view of what John Swallow might have done? John Swallow has done something illegal. John Swallow has done nothing illegal, but has done something unethical. John Swallow has done nothing unethical. Don’t know

On a scale of one to five, one being negative and five being positive, please rank your feelings on the following individuals: (1–5, Don’t know) John Swallow Barack Obama Etc.

Note: The following four questions were used to create the political knowledge index.

What percentage vote of the U.S. House and Senate is needed to override a Presidential veto? A majority Two-thirds Three-fourths Unanimous

For how many years is a United States Senator elected—that is how many years are there in one full term of office for a U.S. Senator? 2 years 4 years 6 years 8 years

Which of the following groups claimed responsibility for recent beheadings in Iraq? ISIS UNICEF Al Qaeda Don’t know

Please indicate which office Joe Biden currently holds: U.S. Supreme Court Justice Vice President of the U.S. Speaker of the U.S. House U.S. Senate Majority Leader